TypeError: forward() missing 1 required positional argument: 'events'

[KristofPusztai]

I appreciate this library but the documentation for models is quite lackluster and the jupyter notebook is not enough to re-use this for other, more complex, use cases.

Pretty frustrating that this error is very unclear and no documentation on how exactly inputs should be formatted or how to fix/what is going on... my input data is shaped as follows:

x_train.shape
>>> torch.Size([5720633, 75])

y_train.shape
>>>torch.Size([5720633, 2])

Getting this error when running the lr_finder method:
lrfinder = model.lr_finder(x_train, y_train, batch_size, tolerance=10)

Full Error:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipykernel_517/3901793619.py in <module>
----> 1 lrfinder = model.lr_finder(x_train, y_train, batch_size, tolerance=10)
      2 _ = lr_finder.plot()

~/.local/lib/python3.7/site-packages/torchtuples/base.py in lr_finder(self, input, target, batch_size, lr_min, lr_max, lr_range, n_steps, tolerance, callbacks, verbose, num_workers, shuffle, **kwargs)
    346                 num_workers,
    347                 shuffle,
--> 348                 **kwargs,
    349             )
    350         return lr_finder

~/.local/lib/python3.7/site-packages/pycox/models/cox.py in fit(self, input, target, batch_size, epochs, callbacks, verbose, num_workers, shuffle, metrics, val_data, val_batch_size, **kwargs)
     51         return super().fit(input, target, batch_size, epochs, callbacks, verbose,
     52                            num_workers, shuffle, metrics, val_data, val_batch_size,
---> 53                            **kwargs)
     54 
     55     def _compute_baseline_hazards(self, input, df, max_duration, batch_size, eval_=True, num_workers=0):

~/.local/lib/python3.7/site-packages/torchtuples/base.py in fit(self, input, target, batch_size, epochs, callbacks, verbose, num_workers, shuffle, metrics, val_data, val_batch_size, **kwargs)
    292                 val_data, val_batch_size, shuffle=False, num_workers=num_workers, **kwargs
    293             )
--> 294         log = self.fit_dataloader(dataloader, epochs, callbacks, verbose, metrics, val_dataloader)
    295         return log
    296 

~/.local/lib/python3.7/site-packages/torchtuples/base.py in fit_dataloader(self, dataloader, epochs, callbacks, verbose, metrics, val_dataloader)
    234                     break
    235                 self.optimizer.zero_grad()
--> 236                 self.batch_metrics = self.compute_metrics(data, self.metrics)
    237                 self.batch_loss = self.batch_metrics["loss"]
    238                 self.batch_loss.backward()

~/.local/lib/python3.7/site-packages/torchtuples/base.py in compute_metrics(self, data, metrics)
    180         out = self.net(*input)
    181         out = tuplefy(out)
--> 182         return {name: metric(*out, *target) for name, metric in metrics.items()}
    183 
    184     def _setup_metrics(self, metrics=None):

~/.local/lib/python3.7/site-packages/torchtuples/base.py in <dictcomp>(.0)
    180         out = self.net(*input)
    181         out = tuplefy(out)
--> 182         return {name: metric(*out, *target) for name, metric in metrics.items()}
    183 
    184     def _setup_metrics(self, metrics=None):

~/.local/lib/python3.7/site-packages/torch/nn/modules/module.py in _call_impl(self, *input, **kwargs)
   1192         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks
   1193                 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1194             return forward_call(*input, **kwargs)
   1195         # Do not call functions when jit is used
   1196         full_backward_hooks, non_full_backward_hooks = [], []

TypeError: forward() missing 1 required positional argument: 'events'